Increase input character limit and output token limit for Kagi Assistant

dreifach

yiwei-1 Got it. Might it be possible to add a message alerting the user to when the Assistant begins/is truncating queries/context? Or would this be unnecessary after the fair use policy's implementation?

yiwei-1

dreifach

We will gauge the usage after implementing the fair use policy to see how often people would hit the context limit/burn tokens too fast

batuhan

yiwei-1 Can you provide information about fair use?

yiwei-1

batuhan

Providing powerful AI tools requires significant resources. To ensure sustainability, we’re starting to enforce our fair-use policy.

Basically our policy states that you can use AI models based on your plan’s value. For example, a $25 monthly plan allows up to $25 worth of raw token cost across all models (there is a 20% built-in margin that we reserve for providing searches, development and infrastructure for the service).

While most users won’t be affected, those exceeding the generous threshold will have the possibility to renew their subscription cycle instantly. Soon, we’ll introduce credit top-ups for added flexibility. This approach ensures a fair, user-funded model while maintaining quality service and is a simple way to control usage, compared to arbitrary usage limits found in other services.

You can find more details in the announcement blog post: https://blog.kagi.com/assistant-for-all

dreifach

yiwei-1 May I ask what the exact timeline on credit top-ups is? Not too keen on renewing my subscription prematurely.

And what will the process be for non-admin Ultimate members of a family plan? Would they be able to renew the family subscription and/or top up credits themselves? Would the admin be alerted to upcoming additional charges?

And (apologies if this is already documented) what are the credit limits for non-Ultimate members of a family plan? Does their (more limited) usage of the Assistant fall under fair use too, and if so what is the process for renewal/topping up?

Linking the previous two questions, will credits be centralised for family plans both for Ultimate and non-Ultimate members?

Edit (22/04/25): Adding to the above, what are credit amounts equivalent to for Ultimate users on Duo/Family plans? If a single Ultimate user with their own account has $25 worth of credits, does a Duo Ultimate user have $22 worth of credits, and a Family Ultimate user (if all six seats are filled) $18.33?

dreifach

dreifach @yiwei-1 Just checking in on this.

batuhan

yiwei-1 Hasn't the fair use policy been implemented yet?

dreifach

batuhan It has been: Kagi Assistant is now available to all users!

batuhan

dreifach So why wasn't the input output limit increased?

dreifach

batuhan These changes are likely being rolled out over time. Just have to be patient.

batuhan

dreifach Could you please add the Next Release tag then?

dreifach

batuhan Slight confusion here — I am a user and not part of the Kagi team!

dreifach

@yiwei-1 Checking in on this again. In addition to the above, I was also wondering whether credit allowances on annual plans are still calculated on a monthly basis.

yiwei-1

The character limit for the input box (per message) has been raised to 1 million characters a while ago. Fee free to let us know if you still encounter any issue with this.

If you are on an annual plan then your AI credit allowances would also be calculated on annual basis (so technically you can use more than one month of credits in a single month if you are on an annual plan)

dreifach

yiwei-1 Good to hear. Is there a total character limit for single conversations?

Regarding allowances: is there any way for annual plan holders to keep track of their usage? I can imagine a scenario where I burn through credits/tokens in three months without knowing how many I'm using, and then facing the spectre of an annual renewal rather than a cheaper monthly renewal.

Also any word on credit top-ups? On this, for duo/family plans is it the individual Ultimate portion of the plan that would be renewed or the plan as a whole?

And on the nature of credit/token usage for family plans. What are the credit limits for Ultimate members on a duo or family plan? Also $25? And what are the credit limits of non-Ultimate family plan members using smaller/cheaper models included in the Professional plan? I am here referring to my previous query quoted below again for ease.

May I ask what the exact timeline on credit top-ups is? Not too keen on renewing my subscription prematurely.

And what will the process be for non-admin Ultimate members of a family plan? Would they be able to renew the family subscription and/or top up credits themselves? Would the admin be alerted to upcoming additional charges?

And (apologies if this is already documented) what are the credit limits for non-Ultimate members of a family plan? Does their (more limited) usage of the Assistant fall under fair use too, and if so what is the process for renewal/topping up?

Linking the previous two questions, will credits be centralised for family plans both for Ultimate and non-Ultimate members?

Edit (22/04/25): Adding to the above, what are credit amounts equivalent to for Ultimate users on Duo/Family plans? If a single Ultimate user with their own account has $25 worth of credits, does a Duo Ultimate user have $22 worth of credits, and a Family Ultimate user (if all six seats are filled) $18.33?

yiwei-1

dreifach

Usage tracking and credit top-up are both in the works and should be available soon.

For Family/Duo Plan Ultimate members, the monthly AI credit limit should also be $25. Non-Ultimate members should have a $10 monthly limit, the same as individual Professional plan users.

hopeful

How about the "arbitrary restrictions" that "will be removed soon" of 32,000 token sliding window maximum for all conversations and rolling window of 16 messages total (last 14 + initial 2) as documented here: https://help.kagi.com/kagi/ai/assistant.html#usage-limits any update on when these restrictions might be lifted? Thanks a lot!

Luis

hopeful that's also work in progress. We expect to be done with these optimisations in the next few weeks.

ownpacist

I think the rolling window of 7 recent messages is too low for longer conversations even if they stay within the 32k token limit. It would probably be a better idea to inform users that having longer conversations will drain out their AI credits quickly but let them have longer conversations anyway if they so wish to.

« Previous Page Next Page »